How Not To Model The Historical Temperature

by Willis Eschenbach, March 24, 2018 in WUWT


Much has been made of the argument that natural forcings alone are not sufficient to explain the 20th Century temperature variations. Here’s the IPCC on the subject:

(…)

I’m sure you can see the problems with this. The computer model has been optimized to hindcast the past temperature changes using both natural and anthropogenic forcings … so of course, when you pull a random group of forcings out of the inputs, it will perform more poorly.